Many pen-based computer systems provide handwriting recognition software that allows users to enter handwritten data using a pen input device. The handwriting recognition software translates characters in the handwritten data into ASCII characters and displays the ASCII characters to the user. The user may write directly over the displayed ASCII characters with the pen input device to correct any errors in how the software recognized the handwritten input. Unfortunately, the reliability of such handwriting recognition software is currently limited, due in part to the variance in handwriting styles.
Other pen-based computer systems allow users to enter data using on-screen keyboards (i.e., keyboards displayed on a display device). Each displayed key corresponds with a printable character. The displayed keys of the on-screen keyboard are typically configured in the conventional pattern found on most typewriters. To enter data, the user simply taps on a displayed key of the on-screen keyboard with the pen input device. The character represented by the displayed key is automatically entered as input to the computer system. Unfortunately, conventional systems that use on-screen keyboards are difficult to use and prone to error. A user of the conventional on-screen keyboard must enter data one character at a time. In particular, the user must position the pen over a character and then select the character to input the character. Then, the user must position the pen over the next character to be input and select the next character. This is repeated on a per character basis until all of the characters have been input are inputted. As a result, it may take a long period of time to enter data. Moreover, since a user must select each character of the input, the user has a high probability of making an error.